Coagulation factors directly cleave SARS-CoV-2 spike and enhance viral entry

  1. Edward R Kastenhuber
  2. Marisa Mercadante
  3. Benjamin Nilsson-Payant
  4. Jared L Johnson
  5. Javier A Jaimes
  6. Frauke Muecksch
  7. Yiska Weisblum
  8. Yaron Bram
  9. Vasuretha Chandar
  10. Gary R Whittaker
  11. Benjamin R tenOever
  12. Robert E Schwartz
  13. Lewis Cantley  Is a corresponding author
  1. Meyer Cancer Center, Department of Medicine, Weill Cornell Medical College, United States
  2. Department of Microbiology, New York University - Langone Health, United States
  3. Department of Microbiology and Immunology, Cornell University, United States
  4. Laboratory of Retrovirology, The Rockefeller University, United States
  5. Division of Gastroenterology and Hepatology, Department of Medicine, Weill Cornell Medicine, United States
  6. Department of Physiology, Biophysics and Systems Biology, Weill Cornell Medicine, United States
5 figures, 2 tables and 2 additional files

Figures

Figure 1 with 4 supplements
Anticoagulant serine protease inhibitors suppress SARS-CoV-2 entry via inhibition of TMPRSS2.

(A) Peptides derived from two known cleavage sites of SARS-CoV-2 spike were designed with C-terminal fluorophore 5-FAM and N-terminal fluorescence resonance energy transfer (FRET) quencher QXL-520. (B) FDA-approved and investigational serine protease inhibitors were screened by enzymatic assay to inhibit TMPRSS2 cleavage of SARS-CoV-2 S1/S2 peptide substrate. Relative change in fluorescence with respect to DMSO vehicle is shown. Colors indicate the described target of the drugs screened. All drugs screened at 10 µM final concentration. (C) Active form of dabigatran in enzymatic assay for TMPRSS2 inhibition. Relative fluorescence with respect to its corresponding 0.1 N HCl vehicle is shown. (D) Schematic of constructs used to generate SARS-CoV-2 spike-pseudotyped/HIV-1-based particles. (E) Calu3 cells were treated with 10 µM of the indicated drugs for 24 hr prior to infection with HIV-1NL/SARS-CoV-2 pseudovirus. Media was changed at 24 hr post infection and pseudoviral entry was measured by nanoluciferase luminescent signal at 40 hr. (F) Calu3 cells treated with 10 µM of the indicated drugs were monitored for confluence by Incucyte for 40 hr. (G) Pseudoviral entry was measured by nanoluciferase luminescent signal in Calu3 cells treated various concentrations of the indicated drugs for 4 hr prior to infection with SARS-CoV-2 pseudovirus. (H) Caco2 cells were infected with lenti-Cas9-blast and U6-sgRNA-EFS-puro-P2A-tRFP and selected. Neutral controls targeting CD4 (not endogenously expressed) or PHGDH intron 1, two sgRNAs each targeting different regions of ACE2 and TMPRSS2 were included. Cells were subsequently infected with HIV-1NL/SARS-CoV-2 pseudovirus. (I) Caco2 cells co-expressing Cas9 and sgRNAs targeting CD4 (not expressed) or TMPRSS2 were treated with 10 µM camostat, nafamostat, or DMSO vehicle. N = 3, *p < 0.05, two-tailed t-test. Data represented as mean ± SEM.

Figure 1—source data 1

Data and summary statistics for enzymatic and pseudovirus assays.

https://cdn.elifesciences.org/articles/77444/elife-77444-fig1-data1-v2.xlsx
Figure 1—figure supplement 1
Optimization of fluorescence resonance energy transfer (FRET) enzymatic assay.

(A) TMPRSS2 enzymatic assay was performed in AB1 (20 mM Tris-HCl, pH 7.3, 100 mM NaCl, 1 mM EDTA, fresh 1 mM DTT) or AB2 (50 mM Tris-HCl, 150 mM NaCl, pH 8) using 10 µM of either S1/S2 or S2’ peptide substrate. (B) Titration of enzyme concentration was performed (0–1000 nM) with 10 µM S1/S2 substrate. Initial reaction velocity V0 (rate of change in fluorescent signal) each enzyme concentration with 10 µM S1/S2 peptide substrate.

Figure 1—figure supplement 2
Further characterization of HIV-1/SARS-CoV-2 pseudovirus.

A549 cells (which do not express ACE2), A549/ACE2 cells (ectopic ACE2 expression from a lentiviral vector), and Caco2 cells (which express endogenous ACE2 and TMPRSS2) infected with HIV-1NL-based particles pseudotyped with SARS-CoV-2 S or VSV G. N = 3, *p < 0.05, two-tailed t-test. Data represented as mean ± SEM.

Figure 1—figure supplement 3
Further characterization of rVSV∆G/SARS-CoV-2 pseudovirus.

(A) Schematic of constructs used to generate SARS-CoV-2 spike-pseudotyped/VSV-based pseudovirus. (B) Nanoluciferase luminescent signal following addition of rVSV∆G pseudovirus complemented with VSV G, SARS-CoV-2 S, SARS-CoV S, or without complementation with any envelope protein to Calu3 cells. Each pseudovirus was titrated by adding the indicated volume of inoculum, supplemented with fresh media up to 200 µl/well in a 96-well plate. (C–F) Nanoluciferase luminescent signal following infection of (C) Caco2, (D) Calu3, (E) A549/ACE2, or (F) Vero cells with rVSV∆G/SARS-CoV-2 pseudovirus pretreated for 4 hr with 10 µM camostat, nafamostat, dabigatran, or otamixaban, compared with uninfected or infected/untreated cells. Expression status of ACE2 and TMPRSS2 for each cell line is indicated. N = 3. Data represented as mean ± SEM.

Figure 1—figure supplement 4
Evidence of CRISPR knockout efficiency.

(A) Constructs used for CRISPR experiments. (B) Percentage of reads exhibiting wild type, frameshift, or in-frame indels at each locus for the indicated sgRNAs. (C–F) Distribution of reads with deletion or insertion by position within amplicon. (G–J) Distribution of the size of insertions and deletions in each amplicon. Two sgRNAs targeting ACE2 (g1 and g2) and two sgRNAs targeting TMPRSS2 (g1 and g2) were analyzed.

Figure 2 with 1 supplement
Coagulation factors directly cleave SARS-CoV-2 spike.

Initial velocities for the cleavage of SARS-CoV-2 spike S1/S2 and S2’ peptide substrates by (A) TMPRSS2, (B) factor Xa, and (C) thrombin were measured over a range of 0–160 µM substrate. From initial velocity values, enzyme kinetic constants (D) turnover rate Kcat (s–1), (E) affinity constant Km, and (F) specificity constant (Kcat/Km) were obtained for the indicated enzymes with S1/S2 and S2’ peptides. (G–I) Heatmaps depict the initial velocity V0 of cleavage of the indicated peptide substrates and concentrations by (G) TMPRSS2, (H) factor Xa, and (I) thrombin.

Figure 2—source data 1

Data and summary statistics for enzymatic assays.

https://cdn.elifesciences.org/articles/77444/elife-77444-fig2-data1-v2.xlsx
Figure 2—figure supplement 1
Fluorescence resonance energy transfer (FRET) enzymatic assay with modified peptide substrates.

(A–C) Initial reaction velocity with respect to enzyme concentration for peptide substrates of the SARS-CoV-2 spike S1/S2 site (S1S2), with P1 arginine substituted with alanine (S1S2-P1A), or with substitutions in the P3 and P4 position (RR > SQ) with (A) TMPRSS2, (B) factor Xa, or (C) thrombin. (D) List of peptide substrates used in this study. (E) Initial reaction velocity of factor Xa cleavage of SARS-CoV-2 S1/S2 or thrombin-R271 peptide substrates in the presence of 0–100 µM phosphatidylcholine/phosphatidylserine (PC/PS) phospholipid vesicles. (F) Dilute Russell’s viper venom clotting time (dRVVT) assay of pooled normal human plasma, supplemented with 0–100 µM PC/PS phospholipid vesicles. N = 3, *p < 0.05, two-tailed t-test. Data represented as mean ± SEM.

Figure 3 with 2 supplements
Factor Xa and thrombin facilitate SARS-CoV-2 spike-mediated entry.

(A) Calu3 cells were infected with rVSV∆G/SARS-CoV-2 pseudovirus with concomitant treatment with vehicle, 250 nM factor Xa, or 250 nM thrombin. Quantification of the ratio of green fluorescent area to total confluence (4 fields/replicate well, 4 wells/condition). (B) Nanoluciferase luminescent signal was measured following infection with rVSV∆G/SARS-CoV-2 pseudovirus and the addition of either vehicle, factor Xa, or thrombin. The effect of factor Xa on rVSV∆G complemented with either (C) SARS-CoV spike or (D) VSV-G was measured by luminescent signal. Luminescent signal was measured following HIV-1NL/SARS-CoV-2 pseudovirus infection and concomitant treatment with 125–250 nM factor Xa in (E) Calu3 cells, (F) A549/ACE2, and (G) Vero cells following transduction with lentiviral vectors to express GFP or TMPRSS2. Following selection, cells were infected with HIV-1NL/SARS-CoV-2 pseudovirus and concomitantly treated with 125–250 nM factor Xa. Subsequently, nanoluciferase luminescent signal was determined and plotted relative to vehicle-treated control. *p < 0.05, two-tailed t-test. Data represented as mean ± SEM.

Figure 3—source data 1

Data and summary statistics for pseudovirus assays with exogenous proteases.

https://cdn.elifesciences.org/articles/77444/elife-77444-fig3-data1-v2.xlsx
Figure 3—figure supplement 1
Further characterization of coagulation factor-induced SARS-CoV-2 pseudovirus infection.

(A) Representative merged brightfield and green fluorescence images of Calu3 cells without infection or following rVSV∆G/SARS-CoV-2 pseudovirus infection and concomitant treatment with vehicle, 250 nM factor Xa, or 250 nM thrombin (corresponding to Figure 3A). Scale bars represent 300 µm. (B–E) HIV-1NL/SARS-CoV-2 pseudovirus with addition of purified protease or vehicle. Nanoluciferase luminescent signal relative to vehicle-treated control was measured following infection in (B) A549/ACE2 and (C) Vero cells. Cell abundance following protease treatment, relative to vehicle control was determined for (D) A549/ACE2 and (E) Vero cells. N = 3, *p < 0.05, two-tailed t-test. Data represented as mean ± SEM.

Figure 3—figure supplement 2
Assessing relevant protease levels with ex vivo clotting assays.

(A) Factor X activity was determined by dilute Russell’s viper venom clotting time (dRVVT). Pooled normal plasma or FX-deficient plasma with the addition of the indicated concentration of active purified factor Xa were assayed. (B) Prothrombin activity was determined by prothrombin time (PT) via activation with thromboplastin. Pooled normal plasma or prothrombin-deficient plasma with the addition of the indicated concentration of active purified thrombin were assayed. N = 3. Data represented as mean ± SEM.

Figure 4 with 2 supplements
Nafamostat broadly inhibits cleavage of spike peptides by both coagulation factors and transmembrane serine proteases.

Initial velocities for the cleavage of 10 µM SARS-CoV-2 spike S1/S2 (top) and S2’ (bottom) peptide substrates by (A) TMPRSS2, (B) TMPRSS11D/human airway trypsin-like protease (C) factor Xa, and (D) thrombin were measured in the presence of DMSO vehicle, or 10 µM camostat, nafamostat, otamixaban, or dabigatran. The relative activity of (E) factor Xa and (F) thrombin were determined over a range of 0–10 µM of the indicated drugs. Calu3 cells were treated with a range of concentrations of nafamostat with or without addition of 250 nM exogenous factor Xa and infected with (G) rVSV∆G/SARS-CoV-2 pseudovirus or (H) HIV-1NL/SARS-CoV-2 pseudovirus and infectivity was measured by luminescence. N = 3, data represented as mean ± SEM.

Figure 4—source data 1

Data and summary statistics for enzymatic assays to determine the effects of protease inhibitors on host proteases.

https://cdn.elifesciences.org/articles/77444/elife-77444-fig4-data1-v2.xlsx
Figure 4—figure supplement 1
Activity of candidate inhibitors against other proteases.

Initial reaction velocity V0 of furin, TMPRSS4, or neutrophil elastase cleavage of (A) S1/S2 peptide substrate or (B) S2’ peptide substrate, treated with DMSO vehicle, camostat, nafamostat, otamixaban, or dabigatran. 10 µM substrate and 10 µM inhibitor were used.

Figure 4—figure supplement 2
Apixaban rescues effect of factor Xa, related to Figure 4.

Calu3 cells were infected with rVSV∆G/SARS-CoV-2 pseudovirus with addition of protease buffer (left) or factor Xa (right). Cells were treated at the time of infection with DMSO vehicle (black), 1 or 10 µM nafamostat (red), or 1 or 10 µM apixaban (orange). N = 6, *p < 0.05, two-tailed t-test. Data represented as mean ± SEM.

Factor Xa and thrombin increase SARS-CoV-2 infection in lung organoids.

Human pluripotent stem cell (hPSC)-derived lung organoids were infected with SARS-CoV-2. Upon infection, organoids were treated with 170 nM of purified factor Xa or thrombin. (A) Relative level of SARS-CoV-2-N RNA following infection at multiplicity of infection (MOI) = 0.1. (B) Relative level of SARS-CoV-2-E RNA following infection at MOI = 0.1. (C) Relative level of SARS-CoV-2-N RNA following infection at MOI = 0.01. (D) Relative level of SARS-CoV-2-E RNA following infection at MOI = 0.01. N = 16, *p = 0.0144, ***p < 0.0001, two-tailed t-test. Data represented as mean ± SEM.

Figure 5—source data 1

Data and summary statistics for infection assays with exogenous proteases.

https://cdn.elifesciences.org/articles/77444/elife-77444-fig5-data1-v2.xlsx

Tables

Table 1
Kinetics of SARS-CoV-2 spike peptide substrate cleavage.

Kinetic constants obtained from initial velocity studies with varying concentrations of SARS-CoV-2 spike S1/S2 and S2’ peptide substrates. Each estimate is based on seven different concentrations of substrate in 1:2 serial dilution (0–160 µM).

EnzymeSubstrateVmax (µM/s)Kcat (s–1)Km (µM)Ksp (s–1 µM–1)
TMPRSS2S1/S27.71E-046.17E-0324.712.50E-04
TMPRSS2S2'4.60E-043.68E-0360.946.04E-05
Factor XaS1/S22.24E-021.79E-0140.354.43E-03
Factor XaS2'3.04E-052.43E-042.7118.97E-05
ThrombinS1/S26.50E-035.20E-0216.343.18E-03
ThrombinS2'7.34E-045.87E-0313.984.20E-04
Key resources table
Reagent type (species) or resourceDesignationSource or referenceIdentifiersAdditional information
Chemical compound, drugCamostatSelleckCat# S2874
Chemical compound, drugNafamostatSelleckCat# S1386
Chemical compound, drugApixabanMedchem ExpressCat# HY-50667
Chemical compound, drugBetrixabanMedchem ExpressCat# HY-10268
Chemical compound, drugBivalirudin (TFA)Medchem ExpressCat# HY-15664
Chemical compound, drugBoceprevirMedchem ExpressCat# HY-10237
Chemical compound, drugDabigatran etexilateMedchem ExpressCat# HY-10274
Chemical compound, drugEdoxabanMedchem ExpressCat# HY-10264
Chemical compound, drugOtamixabanMedchem ExpressCat# HY-70035
Chemical compound, drugRivaroxabanMedchem ExpressCat# HY-50903
Chemical compound, drugSimeprevirMedchem ExpressCat# HY-10241
Chemical compound, drugSivelestatMedchem ExpressCat# HY-17443
Chemical compound, drugTelaprevirMedchem ExpressCat# HY-10235
Chemical compound, drugDabigatranMedchem ExpressCat# HY-10163
Peptide, recombinant proteinThrombinMillipore SigmaCat# 605195
Peptide, recombinant proteinFactor XaMillipore SigmaCat# 69036
Peptide, recombinant proteinTMPRSS2LSBioCat# LS-G57269
Peptide, recombinant proteinTMPRSS4Aviva System BiologyCat# OPCA0240
Peptide, recombinant proteinFurinThermo Fisher ScientificCat# 1503SE010
Peptide, recombinant proteinNeutrophil elastaseThermo Fisher ScientificCat# 9167SE020
Peptide, recombinant proteinS1/S2AnaspecQXL520-PRRARSVASQ-K(5-FAM)-NH2
Peptide, recombinant proteinS2’AnaspecQXL520-KPSKRSFIED-K(5-FAM)-NH2
Peptide, recombinant proteinTHRB-R271AnaspecQXL520-AIEGRTATSE-K(5-FAM)-NH2
Peptide, recombinant proteinFGB-R44AnaspecQXL520-FFSARGHRPL-K(5-FAM)-NH2
Peptide, recombinant proteinS1/S2-P1AAnaspecQXL520-PRRAASVASQ-K(5-FAM)-NH2
Peptide, recombinant proteinS1/S2-HPNAnaspecQXL520-PSQARSVASQ-K(5-FAM)-NH2
Chemical compound, drugPhosphatidylcholineAvanti Polar LipidsCat# 850375C1,2-Dioleoyl-sn-glycero-3-phosphocholine
Chemical compound, drugphosphatidylserineAvanti Polar LipidsCat# 840035C1,2-Dioleoyl-sn-glycero-3-phospho-L-serine
Cell line (Homo sapiens)Calu3ATCCCat# HTB-55; RRID:CVCL_0609
Cell line (Homo sapiens)A549ATCCCat# CCL-185; RRID:CVCL_0023
Cell line (Homo sapiens)Caco2ATCCCat# HTB-37; RRID:CVCL_0025
Cell line (Chlorocebus sabaeus)VeroLaboratory of Benjamin tenOeverRRID:CVCL_0059
Cell line (Homo sapiens)HEK293TATCCCat# CRL-3216; RRID:CVCL_0063
Recombinant DNA reagentpEGPNThis paper
Recombinant DNA reagentpEGPN-ACE2This paper
Recombinant DNA reagentpEGPN-TMPRSS2This paper
Recombinant DNA reagentLenti-Cas9-blastAddgeneCat# 52962
Recombinant DNA reagentipUSEPRFrancisco Sanchez-Rivera & Scott Lowe
Recombinant DNA reagentCMV-SARS-CoV-2-SSchmidt et al., 2020
Recombinant DNA reagentCCNanoLuc/GFPSchmidt et al., 2020
Recombinant DNA reagentHIV-1NL GagPolSchmidt et al., 2020
Commercial assay or kitNEBuilder master mixNew England BiolabsCat# E2621
Chemical compound, drugXtremeGene9Millipore SigmaCat# 6365787001
Chemical compound, drugPolybreneSanta Cruz BiotechnologyCat# SC-134220
Chemical compound, drugLenti-XTakara BioCat# 631232
Chemical compound, drugG418Sigma-AldrichCat# # G8168
Chemical compound, drugBlasticidinInvivogenCat# ANT-BL-1
Chemical compound, drugPuromycinThermo Fisher ScientificCat# A1113803
Commercial assay or kitCell Lysis BufferPromegaCat# E1531
Commercial assay or kitNanoGlo Luciferase AssayPromegaCat# N1130
Biological sample (Homo sapiens)Normal human plasmaPacific HemostasisCat# 95059–698
Biological sample (Homo sapiens)Factor X-deficient plasmaHaematologic TechnologiesCat# FX-ID
Biological sample (Homo sapiens)Prothrombin-deficient plasmaHaematologic TechnologiesCat# FII-ID
Biological sample (Vipera russelli)Russell’s Viper VenomSigma-AldrichCat# V2501
Strain, strain background (Indiana vesiculovirus)rVSVdG/NG-NanoLucSchmidt et al., 2020
Strain, strain background (SARS-CoV-2)SARS-CoV-2, isolate USA-WA1/2020BEI Resources, NIAID, NIHCat# NR-52281
Sequence-based reagentsgRNA: CD4This studysgRNAGGTGCAATGTAGGAGTCCAA
Sequence-based reagentsgRNA: PHGDH intron1This studysgRNAGGGCGAGAGAGAGAAAATTG
Sequence-based reagentsgRNA: ACE2 g1This studysgRNACACCGCAAAGGCGAGAGATAGTTG
Sequence-based reagentsgRNA: ACE2 g2This studysgRNACACCGACATCTTCATGCCTATGTG
Sequence-based reagentsgRNA: TMPRSS2 g1This studysgRNACACCGCTGGAACGAGAACTACGGG
Sequence-based reagentsgRNA: TMPRSS2 g2This studysgRNACACCGGGGACGGGTAGTACTGAGC
Sequence-based reagentPrimer: CD4-ForwardThis studyPCR primersGATAATGGAGAGATGTTGTTGGTTT
Sequence-based reagentPrimer: CD4- ReverseThis studyPCR primersATGTCCAGGTGCCACTATCCT
Sequence-based reagentPrimer: PHGDH intron 1 – ForwardThis studyPCR primersAAAGCAGAACCTTAGCAAAGAGG
Sequence-based reagentPrimer: PHGDH intron 1 – ReverseThis studyPCR primersGAACTAATTGATACGGGGTGCAT
Sequence-based reagentPrimer: ACE2-g1- ForwardThis studyPCR primersTCCCTACTTTTTGTCGTTATTAGCA
Sequence-based reagentPrimer: ACE2-g1- ReverseThis studyPCR primersGGTGATCCACAGCTAATGTATTGTT
Sequence-based reagentPrimer: ACE2-g2- ForwardThis studyPCR primersTCAAAATGCGATTTCTACAATGTTA
Sequence-based reagentPrimer: ACE2-g2- ReverseThis studyPCR primersTGGGCTTTTCAGATTAAACCATTAT
Sequence-based reagentPrimer: TMPRSS2-g1-ForwardThis studyPCR primersACAAATTCCACCTGCTGGTTATAG
Sequence-based reagentPrimer: TMPRSS2-g1- ReverseThis studyPCR primersACTTCATCCTTCAGGTGTACTCATC
Sequence-based reagentPrimer: TMPRSS2-g2- ForwardThis studyPCR primersCAGGAAATAAACACAAAGAGAATCC
Sequence-based reagentPrimer: TMPRSS2-g2-ReverseThis studyPCR primersACTATGAAAACCATGGATACCAACC
Sequence-based reagentSARS-CoV-2-N-FPCR primersTAATCAGACAAGGAACTGATTA
Sequence-based reagentSARS-CoV-2-N-RPCR primersCGAAGGTGTGACTTCCATG
Sequence-based reagentSARS-CoV-2-E-FPCR primersACAGGTACGTTAATAGTTAATAGCGT
Sequence-based reagentSARS-CoV-2-E-RPCR primersATATTGCAGCAGTACGCACACA
Sequence-based reagentHuman 18S-FPCR primersGGCCCTGTAATTGGAATGAGTC
Sequence-based reagentHuman 18S-RPCR primersCCAAGATCCAACTACGAGCTT
Software, algorithmPrism 9GraphPad Software

Additional files

Transparent reporting form
https://cdn.elifesciences.org/articles/77444/elife-77444-transrepform1-v2.docx
Supplementary file 1

Graphical abstract: Positive feedback in SARS-CoV-2 infection and coagulopathy.

In this study, we investigated the role of coagulation factors in SARS-CoV-2 infection. Hyperactivated coagulation is a feature of COVID-19 pathology. Coagulation factors, including factor Xa and thrombin, can cleave SARS-CoV-2 spike. This activity can exacerbate infection by enhancing viral entry. Lastly, we show that a subset of protease inhibitors with anticoagulant properties, such as nafamostat, also have the potential to block host-mediated spike activation by multiple human proteases.

https://cdn.elifesciences.org/articles/77444/elife-77444-supp1-v2.zip

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Edward R Kastenhuber
  2. Marisa Mercadante
  3. Benjamin Nilsson-Payant
  4. Jared L Johnson
  5. Javier A Jaimes
  6. Frauke Muecksch
  7. Yiska Weisblum
  8. Yaron Bram
  9. Vasuretha Chandar
  10. Gary R Whittaker
  11. Benjamin R tenOever
  12. Robert E Schwartz
  13. Lewis Cantley
(2022)
Coagulation factors directly cleave SARS-CoV-2 spike and enhance viral entry
eLife 11:e77444.
https://doi.org/10.7554/eLife.77444